Simulation Engineer

Hammerhead AI•Redwood City, CA

About The Position

As an AI/ML Simulation Engineer, you will build the virtual world where our intelligence is born. You will be responsible for creating and scaling high-fidelity digital twins of physical data center environments. These simulations are the critical training ground for our Orchestrated RL Control Agents (ORCA). Reporting to the CTO, you will model the complex interplay of power, thermal dynamics, and computation. Your work will enable our RL Engineers to safely and rapidly develop control agents that can be deployed with confidence into live, mission-critical facilities.

Requirements

Digital Twin Experience: 3+ years of professional experience in developing digital twins or high-fidelity simulations for complex physical systems (e.g., in energy, aerospace, manufacturing, or robotics).
Strong Programming Skills: Proficiency in Python, Go and/or C++ and experience with simulation frameworks or libraries.
Physics-Based Modeling: Strong understanding of first-principles modeling, with experience capturing the dynamics of physical systems (thermal, electrical, or mechanical).
ML/AI Exposure: Familiarity with the lifecycle of machine learning models and experience creating environments for training and testing AI agents.
Educational Background: MS or PhD in a relevant engineering discipline, Computer Science, Math or Physics.
Problem Solver: A practical mindset, able to abstract complex physical interactions into computationally efficient models and troubleshoot discrepancies between simulation and reality.

Responsibilities

Digital Twin Development: Architect and build high-fidelity, physics-based simulations of data center components, including cooling systems, power distribution units, and server racks.
Simulation Platform Integration: Integrate individual asset models into a comprehensive, scalable simulation platform that represents entire data center environments.
Model Validation: Develop methodologies to validate simulation accuracy against real-world operational data, ensuring our digital twins faithfully represent physical reality.
AI Training Environments: Create and maintain the infrastructure and APIs that allow Reinforcement Learning engineers to train and evaluate control agents at scale within the simulation.
Performance Optimization: Ensure the simulation platform is fast, scalable, and efficient to accelerate the AI/ML development lifecycle.